INVISIBLE HAND Econ Journal Watch Volume 6, Number 2 May 2009, pp 239-263 Editor’s Note on Style: This article follows Professor Kennedy’s British style in spelling and quotation marks. Italics are used to distinguish Adam Smith’s words in the text, though not in block quotations. Adam Smith and the Invisible 1 Hand: From Metaphor to Myth Gavin Kennedy2 ABSTR A C T It almost seems as if Adam Smith, who only intends to write an economic Inquiry into the Nature and Causes of the Wealth of Nations, is led by an invisible hand to promote and end which was not part of his intentions: the writing of a Gothic Novel. -Stefan Andriopoulos (1999, 753) References to an ‘invisible hand’ that link it to Adam Smith are ubiquitous in books and articles from scholarly and media sources. This is strange because Adam Smith did not credit the invisible hand metaphor with the importance that authors, from the mid-20th century onwards, give to it. In this paper I discuss what Adam Smith most probably meant by his use of the ‘invisible hand’ metaphor, which is quite different from what has become its modern meaning. Among recent contributors, William Grampp (2000; cf Minowitz 2004) identified nine different meanings given to the invisible hand (ten, including his own, strange one) in modern literature. Warren Samuels has published an authoritative account and analysis of the way the invisible hand metaphor has been used by modern economists, and, therefore, I have not addressed the details of this almost wholly 20th century phenomenon. Emma Rothschild gave a detailed exposition of the invisible hand and what Smith meant by it – a ‘mildly ironic joke’ in her considered view (Rothschild, 1994, 2001). 1 This paper was originally presented to the 34th Annual Meeting of the History of Economics Society Annual Meeting, George Mason University, June, 2007, and The Journal of the History of Economic Thought, 40th Anniversary Conference, University of Edinburgh, 3 September 2008. 2 Emeritus Professor, Heriot-Watt University, Edinburgh, Scotland, UK EH14 4AS. 239 VOLUME 6, NUMBER 2, MAY 2009 KENNEDY I shall argue that Smith had no ‘theory’ of invisible hands and that he showed no inclination to treat it as anything more than an isolated, though well-known, 18th- century literary metaphor. Significantly, and contrary to the assertions of the modern consensus, he gave the invisible hand no role in his theory of competitive markets in Books I and II of Wealth of Nations. Such roles given to it since the 1950s rely solely on assertions and interpolations by modern economists, which are not supported by Smith’s texts. The ‘invisible hand’ appears once in Smith’s History of Astronomy (Smith 1980, 95) referring to pagan and heathen superstitions about the existence of the Roman god, Jupiter; once in Moral Sentiments (184-185) referring to feudal lords divvying up their produce among their retainers and tenants in roughly the same proportions as would be distributed if the land had been divided equally; and once in Wealth of Nations (456), referring to degrees of caution about the risks associated with distant trade with the British colonies in North America, which incentivised some, but not all, merchants to act circumspectly in their preference for domestic projects, thereby unintentionally benefiting the domestic economy. This is only three times in over a million words published in his surviving essays and books, written between c.1744 and 1790. Yet, three leading economists lauded the invisible hand metaphor and gave it uncalled-for prestige in support of their own interpretations of appropriate economic policies. They described it variously as: • ‘The profoundest observation of Smith’ … ‘the system works behind the backs of the participants; the directing hand is invisible’ (Arrow 1987, 71); • ‘surely the most important contribution [of] economic thought’ (Arrow & Hahn 1971, 1); • ‘one of the great ideas of history and one of the most influential’ (Tobin 1992, 117). • ‘The argument of Adam Smith (1776) that free markets lead to efficient outcomes, “as if by and invisible hand,” has play a central role in these debate’ (Stiglitz 2002, 460, 477; see also Stiglitz 2000, 1448, 1457). Modern benign invisible hand explanations from the second half of the 20th century elevated the metaphor into ‘principles’, ‘theories’ and ‘paradigms’ of markets, which do not correspond to anything written by Smith and neither do they explain anything. How did Smith’s casual metaphor achieve such high status when neither he nor readers, up to the late 19th century, appear to have taken much notice of it? Nor was it mentioned by Governor T. Pownall (1776). The invisible hand was not mentioned by Dugald Stewart (1793), Thomas Malthus (1798), David Ricardo (1817), J. S. Mill (1849), Karl Marx (1887), or J. R. McCulloch (1863). The ‘invisible-hand’ paragraph was paraphrased by H. T. Buckle (1885 vol i, 218-219), but he ignored the metaphor. August Onken (1874) quoted the invisible hand without comment. The metaphor of ‘an invisible hand’ had little if any significance for Smith, EcON JOURNAL WAtcH 240 INVISIBLE HAND and certainly was not his ‘greatest idea’ (Tobin 1992; cf Schneider 1979, 51) nor did the metaphor make ‘theoretical social science itself possible’ (Vaughn 1983, 997). Frederick Hayek, the original author of Karen Vaughn’s ascription of the role of the invisible hand as the gateway to making social science possible, took Smith’s ‘borrowed’ metaphor to be a first approximation of his themes of ‘spontaneous order’ (Hayek 1960). Murray Rothbard cites: Chuang-tzu (369-286 BC): “Good order results spontaneously when things are let alone” (Rothbard 1990). Smith’s identification of the processes associated with the unintended consequences of individual actions in such diverse phenomena as language, money, moral sentiments, exchange and markets (Otteson, 2001), across social experience, are usefully judged to be an early recognition of evolutionary ‘emergent order.’ This phrase is more helpful than spontaneous, or instantly generated, order because it takes many self-correcting trials over long periods for a workable order to emerge as an accepted norm, whereas ‘spontaneous order’ suggests sudden change (an outcome) rather than evolving many steps at a time, most of them in error (a process). Complex systems like language and markets do not emerge suddenly or spontaneously and most certainly do not emerge by design; a long maturation period is required to bring them to term and great and persistent effort. Two people trying to communicate in the absence of a common language make many mistakes as they experiment with different combinations of gestures, grimaces and strings of word sounds to make even simple meanings mutually understood, and, as important, to make them understood by third parties; hominids striking stones to make meat- cutters or axes would miss-strike and regularly break near-finished hand tools forcing them to start again, before they achieved proficiency and created workable tools replicable by others (Bradley and Edmonds 2005). Those who did not master the art lost the opportunity to improve their chances of surviving long enough to breed and bring children to maturity. Time, except for the individual, was not at a premium; the emergent order of successful stone-knapping technology lasted a million years. Long periods of individual experiments, including many dead-ends, and even longer periods of changeless low-level technologies right up to the 18th-century and beyond suggest the absence of an invisible hand operating on the far-slower time scales than the two examples given in Moral Sentiments and Wealth of Nations. One of the problems with associating the concepts of spontaneous order and invisible hands is to conclude that the outcomes of individual actions are necessarily always, or mostly, benign. Robert Nozick identifies 16 examples of ‘Invisible-hand explanations’ (Nozick 1974: 20-21), covering evolutionary theory, ecology, race, religion, genetics, IQ, pricing, equilibria in markets, crime, trade, managerial incompetence, and economic theories, and not all of those he cited have benign outcomes. Karen Vaughn, however, accepts correctly that ‘one could easily imagine a spontaneous order in which people were led as if by an invisible hand to promote a perverse and unpleasant end’ (emphasis added). She comments that ‘the desirability of the order that emerges as the unintended consequences of human action depends ultimately on the kind of rules and institutions within which human beings act, and the real alternatives they face’ (Vaughn 1983). Vaughn correctly undid the connection 241 VOLUME 6, NUMBER 2, MAY 2009 KENNEDY between the existence and the goodness of an invisible hand, but her insight has not been picked up by many modern economists. Parenthetically, in Smith’s usage, the ‘invisible hand’ did not act ‘as if’ it led people; the metaphor in his two examples were definitely ‘led by an invisible hand’ (TMS, 184; WN, 456). The words, ‘as if ’, would act as a softener of the stronger imperative, ‘was led by’, and change its character. The invisible hand is thought of as Smith’s metaphor, but he didn’t invent it. Scholars report many early literary references to ‘invisible hands’ and I have drawn on their work for what follows (Rothschild 2001, 116-56; Force 2003, 69-71; Buchan 2006, 1; Andriopoulos 1999, 739n-758), showing substantial prior use of the metaphor before Smith and with whose work he was familiar (he had many of their books in his library) (Bonar 1966). • Homer (Iliad, 720 BC): ‘And from behind Zeus thrust him [Hector] on with exceeding mighty hand’. Smith had several copies in his library. • Horace: Fulminantis manus Jovis’ (The mighty hand of thundering Jove) Odes 3.3.6, ‘which Smith knew well’ (Force 2003, 70). • Ovid of Caeneus at Troy: ‘twisted and plied his invisible hand, inflicting wound within wound’ (Bonar 1966, 125).